The invention relates to a method for locating faults in a complex system and a device for implementing this method.
In the context of the present invention, a complex system is understood to be a system comprising a plurality of components, notably electronic components, and computers having many interconnections. Such complex systems exist in extremely varied fields of activity, for example in industry, in automobile engineering or in aviation.
In a complex system of this type, notably an aircraft, there are generally monitoring devices capable of determining the existence of a fault or a malfunction in said system, notably by checking the consistency of information transmitted by different components of said system, for example for safety reasons.
However, because of the many interconnections, defective operation of one component of said system has direct or indirect effects on other components of the system so that a single fault can be the source of a multitude of information transmitted by different components and each capable of being considered as transmitted by a defective component.
Thus, although it is generally possible to determine more or less easily the existence of a fault in such a complex system, it is often difficult or even impossible to locate such a fault.
This difficulty stems notably from the fact that it is impossible to store all the available information transmitted by the different components of the system, for example during an operating cycle of the system and capable of being used, for example by a maintenance operator, to locate a fault. In addition, even if such storage were technically feasible the mass of information stored would then be too great for a maintenance operator to be able to deduce any fault location from it.